DMSIG - On Diversity, Complexity, and Regularization in Ensemble Models, January 24, 2011
Date: Monday January 24, 2011; 6:30 pm 6:30 - 9:00 pm (6:30 - 7:00 networking & snacks; 7:00 - 7:10 announcements; 7:10+ presentation, Q&A)
Cost: Free and open to all who wish to attend, but membership is only $20/year. Anyone may join our mailing list at no charge, and receive announcements of upcoming events.
Speakers: Giovanni Seni, PhD
Title: On Diversity, Complexity, and Regularization in Ensemble Models
The discovery of ensemble methods is one of the most influential developments in Data Mining and Machine Learning in the past decade. These methods combine multiple models into a single predictive system that is more accurate than even the best of its components. The use of ensemble methods can provide a critical boost to existing systems addressing the hardest of industrial challenges - from investment timing to drug discovery, from fraud detection to recommendation systems - where predictive accuracy is vital. This talk, based on a recently published book by the speaker, offers a concise introduction to this breakthrough topic. After a sketch of the major concerns in predictive learning, the talk will give an overview of regularization, a key concept driving the superior performance of modern ensemble algorithms. It then takes a shortcut into the heart of the popular tree-based ensemble creation strategies using recent developments from the frontiers of statistics, where research efforts are now focused to explain and harness the mysteries of ensembles.
Giovanni Seni is a Senior Scientist with Elder Research, Inc. (ERI) and directs ERI's Western office. As an active data mining practitioner in Silicon Valley, he has over 15 years R&D experience in statistical pattern recognition, data mining, and human-computer interaction applications. He has been a member of the technical staff at large technology companies, and a contributor at smaller organizations. He holds five US patents and has published over twenty conference and journal articles. His book with John Elder, "Ensemble Methods in Data Mining - Improving accuracy through combining predictions", was published in February 2010 by Morgan & Claypool. Giovanni is also an adjunct faculty at the Computer Engineering Department of Santa Clara University, where he teaches an Introduction to Pattern Recognition and Data Mining class.